[Web CV] Deploy MobileNet for the web
＜实践复现此项目＞
课程：https://www.coursera.org/learn/browser-based-models-tensorflow/home/welcome
代码：https://github.com/lmoroney/dlaicourse
参考：[MobileNet] Train MobilenNet in Keras & TF Hub
Jeff 2020: 直接跳到"项目复现"即可。
基础练习
一、the first training.
Build a model, but not load a model, and then training.
<html> <head></head> <script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@latest"></script> <script lang="js"> async function doTraining(model){ const history = await model.fit(xs, ys, { epochs: 500, callbacks:{ onEpochEnd: async(epoch, logs) =>{ console.log("Epoch:" + epoch + " Loss:" + logs.loss); } } }); } const model = tf.sequential(); model.add(tf.layers.dense({units: 1, inputShape: [1]})); model.compile({loss:'meanSquaredError', optimizer:'sgd'}); model.summary(); const xs = tf.tensor2d([-1.0, 0.0, 1.0, 2.0, 3.0, 4.0], [6, 1]); const ys = tf.tensor2d([-3.0, -1.0, 2.0, 3.0, 5.0, 7.0], [6, 1]); doTraining(model).then(() => { alert(model.predict(tf.tensor2d([10], [1,1]))); }); </script> <body> <h1>First HTML Page</h1> </body> </html>
二、训练可视化
三、Image Classification Using MobileNet
When you launch the mobilenet.html file in the Chrome browser make sure to open the Developer Tools to see the output in the Console.
<html> <head> <script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@latest"> </script> <script src="https://cdn.jsdelivr.net/npm/@tensorflow-models/mobilenet@1.0.0"> </script> </head> <body> <img id="img" src="coffee.jpg"></img> <div id="output" style="font-family:courier;font-size:24px;height=300px"></div> </body> <script> const img = document.getElementById('img'); const outp = document.getElementById('output'); mobilenet.load().then(model => { model.classify(img).then(predictions => { console.log(predictions); for(var i = 0; i<predictions.length; i++){ outp.innerHTML += "<br/>" + predictions[i].className + " : " + predictions[i].probability; } }); }); </script> </html>
自定义模型
线性模型
一、简单的线性模型
模型定义
model = tf.keras.models.Sequential([ # Note the input shape is the desired size of the image 150x150 with 3 bytes color tf.keras.layers.Conv2D(16, (3,3), activation='relu', input_shape=(150, 150, 3)), tf.keras.layers.MaxPooling2D(2,2), tf.keras.layers.Conv2D(32, (3,3), activation='relu'), tf.keras.layers.MaxPooling2D(2,2), tf.keras.layers.Conv2D(64, (3,3), activation='relu'), tf.keras.layers.MaxPooling2D(2,2),
# --------------------------------------------------- # Flatten the results to feed into a DNN tf.keras.layers.Flatten(), # 512 neuron hidden layer tf.keras.layers.Dense(512, activation='relu'), # Only 1 output neuron. It will contain a value from 0-1 where 0 for 1 class ('cats') and 1 for the other ('dogs') tf.keras.layers.Dense(1, activation='sigmoid') ])
确实没有MobileNet的收敛速度快．
Train for 100 steps, validate for 50 steps Epoch 1/15 100/100 - 19s - loss: 0.7329 - accuracy: 0.5675 - val_loss: 0.6213 - val_accuracy: 0.6590 Epoch 2/15 100/100 - 18s - loss: 0.6094 - accuracy: 0.6920 - val_loss: 0.5722 - val_accuracy: 0.7210 Epoch 3/15 100/100 - 18s - loss: 0.5236 - accuracy: 0.7470 - val_loss: 0.5856 - val_accuracy: 0.7050 Epoch 4/15 100/100 - 19s - loss: 0.4484 - accuracy: 0.7925 - val_loss: 0.7536 - val_accuracy: 0.6630 Epoch 5/15 100/100 - 20s - loss: 0.3529 - accuracy: 0.8350 - val_loss: 0.5920 - val_accuracy: 0.7170 Epoch 6/15 100/100 - 19s - loss: 0.2833 - accuracy: 0.8795 - val_loss: 0.6584 - val_accuracy: 0.7250 Epoch 7/15 100/100 - 19s - loss: 0.2161 - accuracy: 0.9155 - val_loss: 0.8035 - val_accuracy: 0.7250 Epoch 8/15 100/100 - 19s - loss: 0.1417 - accuracy: 0.9420 - val_loss: 0.9094 - val_accuracy: 0.7110 Epoch 9/15 100/100 - 19s - loss: 0.1083 - accuracy: 0.9575 - val_loss: 1.0949 - val_accuracy: 0.7290 Epoch 10/15 100/100 - 20s - loss: 0.0729 - accuracy: 0.9735 - val_loss: 1.2127 - val_accuracy: 0.7310 Epoch 11/15 100/100 - 20s - loss: 0.0610 - accuracy: 0.9785 - val_loss: 1.4155 - val_accuracy: 0.7110 Epoch 12/15 100/100 - 20s - loss: 0.0601 - accuracy: 0.9825 - val_loss: 1.0516 - val_accuracy: 0.6820 Epoch 13/15 100/100 - 20s - loss: 0.0527 - accuracy: 0.9860 - val_loss: 2.0745 - val_accuracy: 0.7280 Epoch 14/15 100/100 - 19s - loss: 0.0826 - accuracy: 0.9825 - val_loss: 1.7811 - val_accuracy: 0.7290 Epoch 15/15 100/100 - 19s - loss: 0.0669 - accuracy: 0.9855 - val_loss: 1.8485 - val_accuracy: 0.7060
二、模型保存
浏览器端直接下载文件．
const saveResult = model.save('downloads://model');
MobileNet模型
一、加载预训练模型
let mobilenet; let model; const webcam = new Webcam(document.getElementById('wc')); const dataset = new RPSDataset();
var rockSamples=0, paperSamples=0, scissorsSamples=0; let isPredicting = false;
//--------------------------------------------------------
async function loadMobilenet() { const mobilenet = await tf.loadLayersModel('https://storage.googleapis.com/tfjs-models/tfjs/mobilenet_v1_0.25_224/model.json'); const layer = mobilenet.getLayer('conv_pw_13_relu'); return tf.model({inputs: mobilenet.inputs, outputs: layer.output}); } async function init(){ await webcam.setup(); mobilenet = await loadMobilenet(); tf.tidy(() => mobilenet.predict(webcam.capture())); } init();
二、准备数据
label = parseInt(elem.id); const img = webcam.capture();
dataset.addExample(mobilenet.predict(img), label);
训练数据的保存：
class RPSDataset {
constructor() { this.labels = [] } addExample(example, label) { if (this.xs == null) { this.xs = tf.keep(example); // The tensor to keep from being disposed. this.labels.push(label); } else { const oldX = this.xs; this.xs = tf.keep(oldX.concat(example, 0)); this.labels.push(label); oldX.dispose(); } }
// for training process. encodeLabels(numClasses) { for (var i = 0; i < this.labels.length; i++) { if (this.ys == null) { this.ys = tf.keep(tf.tidy( () => {return tf.oneHot( tf.tensor1d([this.labels[i]]).toInt(), numClasses)})); } else { const y = tf.tidy( () => {return tf.oneHot( tf.tensor1d([this.labels[i]]).toInt(), numClasses)}); const oldY = this.ys; this.ys = tf.keep(oldY.concat(y, 0)); oldY.dispose(); y.dispose(); } } } }
三、训练数据
async function train() {
dataset.ys = null; dataset.encodeLabels(3); // 给ys赋值完毕
model = tf.sequential({ layers: [ tf.layers.flatten({inputShape: mobilenet.outputs[0].shape.slice(1)}), tf.layers.dense({ units: 100, activation: 'relu'}), tf.layers.dense({ units: 3, activation: 'softmax'}) ] });
const optimizer = tf.train.adam(0.0001); model.compile({optimizer: optimizer, loss: 'categoricalCrossentropy'});
let loss = 0; model.fit(dataset.xs, dataset.ys, { epochs: 10, callbacks: { onBatchEnd: async (batch, logs) => { loss = logs.loss.toFixed(5); console.log('LOSS: ' + loss); } } }); }
项目复现
一、代码下载
项目代码: Retrain MobileNet for the web
二、迁移学习
选择合适的版本：
AttributeError: module 'tensorflow' has no attribute 'app' #34431
Issue: cannot import name 'image_module_info_pb2'
Goto: Creating the TensorFlow Hub pip package using Linux
开始训练：
# cd /path/to/retrain-mobilenet-v1
python -m scripts.retrain -h
# Create folders
mkdir -p tf_files/{bottlenecks,dataset,models,training_summaries}
# Add your dataset
cp -R /path/to/my-dataset/* tf_files/dataset
# Set environment variables
IMAGE_SIZE=128
ARCHITECTURE=mobilenet_0.25_$IMAGE_SIZE
# Start training
python -m scripts.retrain \
--image_dir=tf_files/dataset \
--model_dir=tf_files/models \
--architecture=$ARCHITECTURE \
--output_graph=tf_files/retrained_graph.pb \
--output_labels=tf_files/retrained_labels.txt \
--bottleneck_dir=tf_files/bottlenecks \
--summaries_dir=tf_files/training_summaries/$ARCHITECTURE \
--how_many_training_steps=400 \
--learning_rate=0.001
训练结果：
$ls tf_files/
frozen_graph.pb labels.txt quantized_graph.pb
三、测试模型
测试成功，完美。
$ python scripts/label_image.py --graph=tf_files/retrained_graph.pb --input_width=$IMAGE_SIZE --input_height=$IMAGE_SIZE --image=tf_files/dataset/positive_isDemage/3_roiImg1.jpg
WARNING:tensorflow:From scripts/label_image.py:29: The name tf.GraphDef is deprecated. Please use tf.compat.v1.GraphDef instead. WARNING:tensorflow:From scripts/label_image.py:42: The name tf.read_file is deprecated. Please use tf.io.read_file instead. WARNING:tensorflow:From scripts/label_image.py:56: The name tf.image.resize_bilinear is deprecated. Please use tf.compat.v1.image.resize_bilinear instead. WARNING:tensorflow:From scripts/label_image.py:58: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead. 2020-06-10 13:28:10.983890: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA 2020-06-10 13:28:11.004028: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 1999965000 Hz 2020-06-10 13:28:11.004412: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x557d967b5fe0 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2020-06-10 13:28:11.004438: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version WARNING:tensorflow:From scripts/label_image.py:65: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead. Evaluation time (1-image): 0.068s positive isdemage (score=0.50109) negative isdemage (score=0.49891)
四、Optimize for the web
模型压缩
大小没什么变化；压缩后就能看到quantized的效果。
$ python ./scripts/quantize_graph.py --input=tf_files/retrained_graph.pb --output=tf_files/quantized_graph.pb --output_node_names=final_result --mode=weights_rounded
WARNING:tensorflow:From ./scripts/quantize_graph.py:1191: extract_sub_graph (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version. Instructions for updating: Use `tf.compat.v1.graph_util.extract_sub_graph` W0610 15:01:24.278689 139993801140032 deprecation.py:323] From ./scripts/quantize_graph.py:1191: extract_sub_graph (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version. Instructions for updating: Use `tf.compat.v1.graph_util.extract_sub_graph` WARNING:tensorflow:From ./scripts/quantize_graph.py:1294: FastGFile.__init__ (from tensorflow.python.platform.gfile) is deprecated and will be removed in a future version. Instructions for updating: Use tf.gfile.GFile. W0610 15:01:24.287163 139993801140032 deprecation.py:323] From ./scripts/quantize_graph.py:1294: FastGFile.__init__ (from tensorflow.python.platform.gfile) is deprecated and will be removed in a future version. Instructions for updating: Use tf.gfile.GFile.
jeffrey@unsw-ThinkPad-T490:retrain-mobilenet-v1$ ls tf_files/ -l total 3948 drwxr-xr-x 4 jeffrey jeffrey 4096 Jun 10 13:15 bottlenecks drwxr-xr-x 4 jeffrey jeffrey 4096 Jun 10 13:09 dataset drwxr-xr-x 3 jeffrey jeffrey 4096 Jun 10 13:15 models -rw-r--r-- 1 jeffrey jeffrey 2008026 Jun 10 15:01 quantized_graph.pb -rw-r--r-- 1 jeffrey jeffrey 2008024 Jun 10 13:15 retrained_graph.pb -rw-r--r-- 1 jeffrey jeffrey 36 Jun 10 13:15 retrained_labels.txt drwxr-xr-x 3 jeffrey jeffrey 4096 Jun 10 13:15 training_summaries
$ gzip -k tf_files/retrained_graph.pb tf_files/quantized_graph.pb $ du -h tf_files/*.gz 640K tf_files/quantized_graph.pb.gz 1.8M tf_files/retrained_graph.pb.gz
Convert to TensorFlow.js model
转化为 tfjs模型。
$ tensorflowjs_converter \ --input_format=tf_frozen_model \ --output_node_names=final_result \ tf_files/quantized_graph.pb \ tf_files/web
2020-06-10 15:19:58.882846: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:786] Optimization results for grappler item: graph_to_optimize 2020-06-10 15:19:58.882905: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] debug_stripper: debug_stripper did nothing. time = 0.128ms. 2020-06-10 15:19:58.882913: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] model_pruner: Graph size after: 420 nodes (-140), 446 edges (-140), time = 4.258ms. 2020-06-10 15:19:58.882918: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] constant_folding: Graph size after: 176 nodes (-244), 175 edges (-271), time = 21.271ms. 2020-06-10 15:19:58.882924: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] arithmetic_optimizer: Graph size after: 176 nodes (0), 175 edges (0), time = 4.154ms. 2020-06-10 15:19:58.882929: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] dependency_optimizer: Graph size after: 174 nodes (-2), 173 edges (-2), time = 1.247ms. 2020-06-10 15:19:58.882935: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] model_pruner: Graph size after: 174 nodes (0), 173 edges (0), time = 0.873ms. 2020-06-10 15:19:58.882941: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] constant_folding: Graph size after: 174 nodes (0), 173 edges (0), time = 3.476ms. 2020-06-10 15:19:58.882947: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] arithmetic_optimizer: Graph size after: 174 nodes (0), 173 edges (0), time = 3.773ms. 2020-06-10 15:19:58.882953: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] dependency_optimizer: Graph size after: 174 nodes (0), 173 edges (0), time = 1.043ms. 2020-06-10 15:19:58.882960: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] debug_stripper: debug_stripper did nothing. time = 0.261ms. 2020-06-10 15:19:58.882968: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] model_pruner: Graph size after: 174 nodes (0), 173 edges (0), time = 0.81ms. 2020-06-10 15:19:58.882975: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] constant_folding: Graph size after: 174 nodes (0), 173 edges (0), time = 3.222ms. 2020-06-10 15:19:58.882983: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] arithmetic_optimizer: Graph size after: 174 nodes (0), 173 edges (0), time = 3.627ms. 2020-06-10 15:19:58.882990: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] dependency_optimizer: Graph size after: 174 nodes (0), 173 edges (0), time = 1.035ms. 2020-06-10 15:19:58.882997: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] model_pruner: Graph size after: 174 nodes (0), 173 edges (0), time = 0.742ms. 2020-06-10 15:19:58.883004: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] constant_folding: Graph size after: 174 nodes (0), 173 edges (0), time = 3.115ms. 2020-06-10 15:19:58.883012: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] arithmetic_optimizer: Graph size after: 174 nodes (0), 173 edges (0), time = 3.579ms. 2020-06-10 15:19:58.883019: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] dependency_optimizer: Graph size after: 174 nodes (0), 173 edges (0), time = 1.055ms. 2020-06-10 15:19:58.923034: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:786] Optimization results for grappler item: graph_to_optimize 2020-06-10 15:19:58.923056: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] remapper: Graph size after: 173 nodes (-1), 172 edges (-1), time = 0.691ms. 2020-06-10 15:19:58.923080: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] constant_folding: Graph size after: 173 nodes (0), 172 edges (0), time = 3.373ms. 2020-06-10 15:19:58.923088: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] arithmetic_optimizer: Graph size after: 173 nodes (0), 172 edges (0), time = 3.672ms. 2020-06-10 15:19:58.923095: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] dependency_optimizer: Graph size after: 173 nodes (0), 172 edges (0), time = 0.965ms. 2020-06-10 15:19:58.923102: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] remapper: Graph size after: 173 nodes (0), 172 edges (0), time = 0.734ms. 2020-06-10 15:19:58.923108: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] constant_folding: Graph size after: 173 nodes (0), 172 edges (0), time = 2.933ms. 2020-06-10 15:19:58.923115: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] arithmetic_optimizer: Graph size after: 173 nodes (0), 172 edges (0), time = 3.192ms. 2020-06-10 15:19:58.923122: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:788] dependency_optimizer: Graph size after: 173 nodes (0), 172 edges (0), time = 1.091ms. Writing weight file tf_files/web/model.json...
模型压缩
1.9M压缩为953kb，压缩率大约50%。
$ gzip -k tf_files/web/ group1-shard1of1.bin model.json
jeffrey@unsw-ThinkPad-T490:retrain-mobilenet-v1$ gzip -k tf_files/web/group1-shard1of1.bin jeffrey@unsw-ThinkPad-T490:retrain-mobilenet-v1$ ll tf_files/web/ total 2836 drwxr-xr-x 2 jeffrey jeffrey 4096 Jun 10 15:24 ./ drwxr-xr-x 7 jeffrey jeffrey 4096 Jun 10 15:19 ../ -rw-r--r-- 1 jeffrey jeffrey 1883348 Jun 10 15:19 group1-shard1of1.bin -rw-r--r-- 1 jeffrey jeffrey 953972 Jun 10 15:19 group1-shard1of1.bin.gz -rw-r--r-- 1 jeffrey jeffrey 56654 Jun 10 15:19 model.json
继续压缩：70.3kB
jeffrey@unsw-ThinkPad-T490:differ2_pb$ xxd quantized_graph.pb > file1.hex jeffrey@unsw-ThinkPad-T490:differ2_pb$ xxd quantized_graph_old.pb > file2.hex jeffrey@unsw-ThinkPad-T490:differ2_pb$ diff file1.hex file2.hex > log.diff jeffrey@unsw-ThinkPad-T490:differ2_pb$ ll total 30988 drwxr-xr-x 2 jeffrey jeffrey 4096 Jun 10 15:49 ./ drwxr-xr-x 10 jeffrey jeffrey 4096 Jun 10 15:43 ../ -rw-r--r-- 1 jeffrey jeffrey 8534130 Jun 10 15:49 file1.hex -rw-r--r-- 1 jeffrey jeffrey 8534130 Jun 10 15:49 file2.hex -rw-r--r-- 1 jeffrey jeffrey 10628750 Jun 10 15:49 log.diff -rw-r--r-- 1 jeffrey jeffrey 2008026 Jun 10 15:44 quantized_graph_old.pb -rw-r--r-- 1 jeffrey jeffrey 2008026 Jun 10 15:41 quantized_graph.pb
jeffrey@unsw-ThinkPad-T490:differ1_bin$ ll total 19400 drwxr-xr-x 2 jeffrey jeffrey 4096 Jun 10 15:48 ./ drwxr-xr-x 10 jeffrey jeffrey 4096 Jun 10 15:43 ../ -rw-r--r-- 1 jeffrey jeffrey 8004268 Jun 10 15:47 file1.hex -rw-r--r-- 1 jeffrey jeffrey 8004268 Jun 10 15:48 file2.hex -rw-r--r-- 1 jeffrey jeffrey 1883348 Jun 10 15:42 group1-shard1of1.bin -rw-r--r-- 1 jeffrey jeffrey 1883348 Jun 10 15:19 group1-shard1of1_old.bin -rw-r--r-- 1 jeffrey jeffrey 70316 Jun 10 15:48 log.diff
Add labels
# Install 'jq' once with homebrew (https://brew.sh/) brew install jq # Create JSON file from newline-delimited text file cat tf_files/retrained_labels.txt | jq -Rsc '. / "\n" - [""]' > tf_files/web/labels.json cat tf_files/web/labels.json
五、Deploy on the web
tensorflowjs_model.pb, weights_manifest.json如何来的？
此项目链接web page demo有问题，需看最新的readme，如下：
https://github.com/tensorflow/tfjs-examples/tree/master/mobilenet [可用]
https://github.com/tensorflow/tfjs/tree/master/tfjs-converter/demo/mobilenet [命令参考]
$ npm install
$ npm run build > tfjs-examples-mobilenet@0.1.0 build /home/jeffrey/Desktop/tfjs-examples-master/mobilenet > cross-env NODE_ENV=production parcel build index.html --no-minify --public-url ./ ✨ Built in 1.75s. dist/mobilenet.1fda360a.map ⚠️ 5.1 MB 472ms dist/mobilenet.ea2308db.js ⚠️ 1.34 MB 1.07s dist/tfjs-examples.c3b5157d.css 280.77 KB 7ms dist/cat.d8f4b943.jpg 40.08 KB 8ms dist/index.html 2.5 KB 3ms
jeff@unsw-ThinkPad-T490:mobilenet$ ls cat.jpg dist imagenet_classes.js index.html index.js node_modules package.json README.md yarn.lock
我们就可以像打开静态网页一样打开我们完成的项目。
打包生成的dist文件夹，其中index.html为入口文件。
六、布置静态网页
Ref: Ubuntu 18.04系统中使用Apache搭建简单的web服务器
安装命令：
sudo apt install apache2 -y
启动 Web 服务器：
jeffrey@unsw-ThinkPad-T490:~$ systemctl status apache2 systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libcryptsetup.so.12) systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libblkid.so.1) ● apache2.service - The Apache HTTP Server Loaded: loaded (/lib/systemd/system/apache2.service; enabled; vendor preset: Drop-In: /lib/systemd/system/apache2.service.d └─apache2-systemd.conf Active: active (running) since Thu 2020-06-11 09:54:41 AEST; 10s ago Main PID: 31523 (apache2) Tasks: 55 (limit: 4915) CGroup: /system.slice/apache2.service ├─31523 /usr/sbin/apache2 -k start ├─31524 /usr/sbin/apache2 -k start └─31525 /usr/sbin/apache2 -k start jeffrey@unsw-ThinkPad-T490:~$ jeffrey@unsw-ThinkPad-T490:~$ jeffrey@unsw-ThinkPad-T490:~$ /etc/init.d/apache2 start [....] Starting apache2 (via systemctl): apache2.service/bin/systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libcryptsetup.so.12) /bin/systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libblkid.so.1) /bin/systemd-tty-ask-password-agent: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libcryptsetup.so.12) /bin/systemd-tty-ask-password-agent: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libblkid.so.1) . ok jeffrey@unsw-ThinkPad-T490:~$ jeffrey@unsw-ThinkPad-T490:~$ jeffrey@unsw-ThinkPad-T490:~$ /etc/init.d/apache2 stop [....] Stopping apache2 (via systemctl): apache2.service/bin/systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libcryptsetup.so.12) /bin/systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libblkid.so.1) /bin/systemd-tty-ask-password-agent: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libcryptsetup.so.12) /bin/systemd-tty-ask-password-agent: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libblkid.so.1) . ok jeffrey@unsw-ThinkPad-T490:~$ jeffrey@unsw-ThinkPad-T490:~$ jeffrey@unsw-ThinkPad-T490:~$ /etc/init.d/apache2 restart [....] Restarting apache2 (via systemctl): apache2.service/bin/systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libcryptsetup.so.12) /bin/systemctl: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libblkid.so.1) /bin/systemd-tty-ask-password-agent: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libcryptsetup.so.12) /bin/systemd-tty-ask-password-agent: /usr/local/anaconda3/lib/libuuid.so.1: no version information available (required by /lib/x86_64-linux-gnu/libblkid.so.1) . ok
遇到CORS问题：
$ google-chrome --disable-web-security
修改 Nginx 后台配置文件
Nginx命令参考：如何在Ubuntu 18.04上安装Nginx
现在您的Web服务器已启动并运行，让我们来回顾一些基本的管理命令。
要停止您的Web服务器，请键入： sudo systemctl stop nginx 停止时要启动Web服务器，请输入： sudo systemctl start nginx 要停止并再次启动服务，请键入： sudo systemctl restart nginx 如果您只是简单地进行配置更改，Nginx通常可以重新加载而不会丢失连接。 为此，请输入： sudo systemctl reload nginx 默认情况下，Nginx配置为在服务器引导时自动启动。 如果这不是您想要的，可以通过输入以下命令来禁用此行为： sudo systemctl disable nginx 要重新启用服务以在启动时启动，您可以键入： sudo systemctl enable nginx
Nginx配置实例
实例一：允许example.com的应用在www.example2.com上跨域提取数据
在nginx.conf里找到server项,并在里面添加如下配置
1
|
location /{
|
如果需要允许来自任何域的访问，可以这样配置
1
|
add_header Access-Control-Allow-Origin *;
|
End.
【推荐】国内首个AI IDE，深度理解中文开发场景，立即下载体验Trae
【推荐】编程新体验，更懂你的AI，立即体验豆包MarsCode编程助手
【推荐】抖音旗下AI助手豆包，你的智能百科全书，全免费不限次数
【推荐】轻量又高性能的 SSH 工具 IShell：AI 加持，快人一步
· AI与.NET技术实操系列：向量存储与相似性搜索在 .NET 中的实现
· 基于Microsoft.Extensions.AI核心库实现RAG应用
· Linux系列：如何用heaptrack跟踪.NET程序的非托管内存泄露
· 开发者必知的日志记录最佳实践
· SQL Server 2025 AI相关能力初探
· 震惊！C++程序真的从main开始吗？99%的程序员都答错了
· 【硬核科普】Trae如何「偷看」你的代码？零基础破解AI编程运行原理
· 单元测试从入门到精通
· 上周热点回顾（3.3-3.9）
· winform 绘制太阳，地球，月球 运作规律